Malaria caused by the protozoan parasite Plasmodium falciparum infects about a million people each day and causes 5000 deaths, mostly in children. Drug resistance in the parasite and insecticide resistance in the vector mosquitoes have combined to make malaria one of the principal public health issues in the developing world. More basic research has been called for in the hope that greater understanding will lead to new approaches for treatment and prevention. Population genetics can contribute to this effort because there is evidence that all extant P. falciparum share a recent common ancestor. This implies the most of the genetic variation present in the organism has been maintained by directional or balancing selection. Genes that are exceptionally polymorphic may therefore offer novel targets for drug design. The evidence for a recent common ancestor is, however, mixed and difficult to interpret. The ambiguities result from the fact that the biology and population history of P. falciparum is considerably more complex than originally thought. The age of the common ancestor has therefore become a convenient proxy for addressing these more fundamental issues of parasite biology. To resolve some of these issues, we will study single-nucleotide polymorphisms (SNPs) in non-coding regions of DNA across four chromosomes to determine whether the distribution is consistent with a single coalescent. We will also assess the effects of choice of reference strains by comparing isolates based on geographical diversity with those based on microsatellite diversity. Types and levels of SNP variation will be studied in-depth in recently archived field isolates from Africa, South America and Asia in order to determine the extent to which cultured isolates represent the full range of genetic variation present in the species. We will study genetic differences between P. falciparum and its closest known relative, the chimpanzee parasite P. reichenowi, using the resources of the P. reichenowi Genome Project. These studies will allow improved estimates of the mutation rate for synonymous sites and the first estimates for various classes of non-coding DNA, and will address the key issue whether the mutation rate is elevated in microsatellite repeats. Comparisons of coding regions will allow analyses of polymorphism and divergence to test whether individual genes are subject to selection. We will use the genome sequence of P. falciparum to carry out bioinformatics analyses of gene duplications and codon usage. Gene paralogs will be studied to estimate rates and patterns of sequence divergence especially in non-coding DNA. Codon usage will be studied in relation to gene expression to test the hypothesis that selection for optimal codon usage leads to a bias in the density of non-synonymous SNPs. [unreadable] [unreadable]